CN106033620A - Point clouds model restoration method, point clouds model restoration device and calculation device - Google Patents
Point clouds model restoration method, point clouds model restoration device and calculation device Download PDFInfo
- Publication number
- CN106033620A CN106033620A CN201510111559.6A CN201510111559A CN106033620A CN 106033620 A CN106033620 A CN 106033620A CN 201510111559 A CN201510111559 A CN 201510111559A CN 106033620 A CN106033620 A CN 106033620A
- Authority
- CN
- China
- Prior art keywords
- cloud model
- point cloud
- point
- neighborhood
- scattered points
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Landscapes
- Processing Or Creating Images (AREA)
Abstract
The invention discloses a point clouds model restoration method, a point clouds model restoration device and a calculation device. The invention discloses a point clouds model restoration method comprises steps of determining a point clouds model to be restored, performing traversal on each of scattered points of the point clouds model, determining a neighborhood corresponding to each scattered point which is performed by traversal, if a neighborhood to be processed with scattered point density lower than a density threshold exists in the neighborhood corresponding to each of the scattered points which is performed by traversal, determining an inserting point in a non-scattered point area in the neighborhood to be processed, and restoring the point clouds model according to the determined inserting point. The point clouds model restoration method, the point clouds model restoration device and the calculation device enable the cavity of the point clouds model to be shrank and improve validity.
Description
Technical field
The present invention relates to threedimensional model processing technology field, be specifically related to a kind of point cloud model restorative procedure,
Device and the equipment of calculating.
Background technology
Point cloud model is that an object (including any entity, such as street flooring, trees etc.) is carried out three
Threedimensional model after dimension modeling, is mainly made up of a scattered points for the object that cloud detecting devices is detected;Mesh
Before in the building process of point cloud model, mainly use LiDar (Light Detection And Ranging,
Optical detection and measurement) equipment carry out object scattered points detection, LiDar equipment is the one of a cloud detecting devices
Kind, the data measured by LiDar equipment are DSM (Digital Surface Model, the digital surface mould of object
Type) discrete point represent, containing the space three-dimensional information of object and laser intensity information in measured data
Deng.
The present inventor finds in research process, clicks on the at random of object at a cloud detecting devices
During row detection, due to blocking or the impact of shadow crevice projection angle of other entities, finally lead to
Cross the object-point cloud model constructed by the scattered points of detected object and there is substantial amounts of cavity, these cavities
Validity by the object-point cloud model constructed by strong influence;As a example by streetscape point cloud model, at car
During carrying the scattered points of LiDar equipment detection building, the laser used due to vehicle-mounted LiDar equipment is worn
Thoroughly the blocking of glass metope and other objects, will cause the building point cloud model of last structure to exist big
The cavity of amount.Fig. 1 shows the schematic diagram of the building point cloud model that there is cavity, it can be seen that institute's structure
The window of the building point cloud model built, glass metope, partly there is substantial amounts of cavity in roof etc., building
The validity of object point cloud model is relatively low.
Therefore, how the point cloud model that there is cavity is repaired, solve the existing structure that inventor finds
There is substantial amounts of cavity in the point cloud model built, the problem that validity is relatively low, becomes those skilled in the art urgently
The problem that need to solve.
Summary of the invention
In view of this, the embodiment of the present invention provides a kind of point cloud model restorative procedure, device and the equipment of calculating,
To repair the object-point cloud model that there is cavity, the object-point cloud model solving existing structure exists
Substantial amounts of cavity, the problem that validity is relatively low.
For achieving the above object, the following technical scheme of embodiment of the present invention offer:
A kind of point cloud model restorative procedure, including:
Determine point cloud model to be repaired;
Travel through each scattered points of described point cloud model, and determine that each scattered points traveled through is corresponding
Neighborhood;
If in the neighborhood that each scattered points traveled through is corresponding, there is scattered points density less than density threshold
Pending neighborhood, then in the region without scattered points of pending neighborhood, determine insertion point;
Described point cloud model is repaired in insertion point determined by according to.
The embodiment of the present invention also provides for a kind of point cloud model prosthetic device, including:
Model determines module, is used for determining point cloud model to be repaired;
Neighborhood determines module, for traveling through each scattered points of described point cloud model, and determines and is traveled through
Neighborhood corresponding to each scattered points;
Insertion point determines module, if in the neighborhood that each scattered points of being traveled through is corresponding, existing scattered
Random dot density is less than the pending neighborhood of density threshold, then in the region without scattered points of pending neighborhood really
Determine insertion point;
Repair module, repairs described point cloud model for insertion point determined by basis.
The embodiment of the present invention also provides for a kind of calculating equipment, including point cloud model prosthetic device described above.
Based on technique scheme, the point cloud model restorative procedure that the embodiment of the present invention provides, treat determining
Repair after point cloud model, can travel through each scattered points of described point cloud model, and determine traveled through every
The neighborhood that one scattered points is corresponding;If in the neighborhood that each scattered points traveled through is corresponding, existing at random
Dot density less than the pending neighborhood of density threshold, then determines in the region without scattered points of pending neighborhood
Insertion point;Described point cloud model is repaired in insertion point determined by according to.The embodiment of the present invention is by traversal point
Each scattered points of cloud model, in the neighborhood that each scattered points is corresponding, finds scattered points density less than density threshold
The pending neighborhood of value, thus determine insertion point in the region without scattered points of pending neighborhood, obtain many
Individual insertion point, is repaired the cavity of point cloud model, can make point cloud model by obtained multiple insertion points
Cavity reduction, improve point cloud model validity.
Accompanying drawing explanation
In order to be illustrated more clearly that the embodiment of the present invention or technical scheme of the prior art, below will be to reality
Execute the required accompanying drawing used in example or description of the prior art to be briefly described, it should be apparent that below,
Accompanying drawing in description is only embodiments of the invention, for those of ordinary skill in the art, not
On the premise of paying creative work, it is also possible to obtain other accompanying drawing according to the accompanying drawing provided.
Fig. 1 is the schematic diagram of the building point cloud model that there is cavity;
The flow chart of the point cloud model restorative procedure that Fig. 2 provides for the embodiment of the present invention;
Another flow chart of the point cloud model restorative procedure that Fig. 3 provides for the embodiment of the present invention;
The method flow diagram of the determination insertion point that Fig. 4 provides for the embodiment of the present invention;
The insertion point that Fig. 5 provides for the embodiment of the present invention determines schematic diagram;
The other method flow chart of the determination insertion point that Fig. 6 provides for the embodiment of the present invention;
Another insertion point that Fig. 7 provides for the embodiment of the present invention determines schematic diagram;
Fig. 8 for the embodiment of the present invention provide according to determined by insertion point repair point cloud model method stream
Cheng Tu;
Another flow chart of the point cloud model restorative procedure that Fig. 9 provides for the embodiment of the present invention;
The another flow chart of the point cloud model restorative procedure that Figure 10 provides for the embodiment of the present invention;
The structured flowchart of the point cloud model prosthetic device that Figure 11 provides for the embodiment of the present invention;
Figure 12 determines the structured flowchart of module for the insertion point that the embodiment of the present invention provides;
Figure 13 determines the structured flowchart of performance element for the insertion point that the embodiment of the present invention provides;
Figure 14 determines another structured flowchart of performance element for the insertion point that the embodiment of the present invention provides;
The structured flowchart of the repair module that Figure 15 provides for the embodiment of the present invention;
Another structured flowchart of the point cloud model prosthetic device that Figure 16 provides for the embodiment of the present invention;
Another structured flowchart of the point cloud model prosthetic device that Figure 17 provides for the embodiment of the present invention;
The hardware block diagram calculating equipment that Figure 18 provides for the embodiment of the present invention.
Detailed description of the invention
Below in conjunction with the accompanying drawing in the embodiment of the present invention, the technical scheme in the embodiment of the present invention is carried out
Clearly and completely describe, it is clear that described embodiment is only a part of embodiment of the present invention, and
It is not all, of embodiment.Based on the embodiment in the present invention, those of ordinary skill in the art are not doing
Go out the every other embodiment obtained under creative work premise, broadly fall into the scope of protection of the invention.
The flow chart of the point cloud model restorative procedure that Fig. 2 provides for the embodiment of the present invention, the method can be applicable to
There is the calculating equipment that point cloud model can carry out data process of data-handling capacity, as smart mobile phone,
The subscriber equipmenies such as PC (personal computer), or the network equipment such as server;Use the embodiment of the present invention
The point cloud model restorative procedure provided, can be carried out the cavity that the point cloud model of arbitrary standalone object exists
Repair;With reference to Fig. 2, the method may include that
Step S100, determine point cloud model to be repaired;
Point cloud model to be repaired can be the arbitrary point cloud model that there is cavity, is not limited to standalone object
Initial point cloud model, it is also possible to be by cavity repair process, but still suffer from the point cloud model in cavity.
Step S110, travel through each scattered points of described point cloud model, and determine each traveled through
The neighborhood that scattered points is corresponding;
Point cloud model has multiple scattered points, and in once traversal, the embodiment of the present invention can travel through a cloud mould
Each scattered points of type, determines the neighborhood that each scattered points traveled through is corresponding;
Optionally, the embodiment of the present invention can set contiguous range, for each scattered points of point cloud model,
Can determine that and comprise traveled through scattered points, and size is to set the neighborhood of contiguous range;Preferably, may be used
Centered by the scattered points traveled through, determine that size is to set the neighborhood of contiguous range;
Further, determined by the shape in field can be rule, as spherical, cuboid, cube
Etc. shape;Scattered points in neighborhood (the neighborhood point that the scattered points that i.e. traveled through is corresponding) can be by KNN
(k-NearestNeighbor, nearest neighbor algorithm) algorithm obtains.
After determining the neighborhood that each scattered points traveled through is corresponding, can draw and described point cloud model
The neighborhood that scattered points quantity is corresponding, a corresponding neighborhood of scattered points.
If in the neighborhood that each scattered points that step S120 is traveled through is corresponding, there is scattered points density low
In the pending neighborhood of density threshold, then in the region without scattered points of pending neighborhood, determine insertion point;
For each neighborhood, the embodiment of the present invention can determine whether that whether the scattered points density in neighborhood is less than setting
Density threshold, if the scattered points density in neighborhood is less than density threshold, then this neighborhood is pending neighborhood,
The insertion point repairing cavity can be determined from this pending neighborhood;The scattered points density of neighborhood can pass through neighborhood
Interior scattered points number, obtains divided by neighborhood area;
Set density threshold and may be considered point cloud model scattered points density in the case of there is not cavity,
The concrete visual practical situations of numerical value sets;
Region without scattered points is the zonule in pending neighborhood, without the area in scattered points region less than waiting to locate
The area of reason neighborhood;Region without scattered points may be considered the district that in pending neighborhood, scattered points distribution is sparse
Territory;Optionally, may be considered point cloud model in the case of there is not cavity without the size in scattered points region,
The area that one scattered points is averagely occupied, the size without scattered points region can regard reality application feelings
Condition sets;
Optionally, insertion point quantity determined by a pending neighborhood is chosen as 1, it is clear that do not limit
In this value.
Described point cloud model is repaired in insertion point determined by step S130, basis.
Optionally, determined by insertion point correct in the case of, insertion point as defined and described point
The morphological characteristic of cloud model matches, then available determined by insertion point repair the cavity of point cloud model.
The point cloud model restorative procedure that the embodiment of the present invention provides, after determining point cloud model to be repaired, can
Travel through each scattered points of described point cloud model, and determine the neighbour that each scattered points traveled through is corresponding
Territory;If in the neighborhood that each scattered points traveled through is corresponding, there is scattered points density less than density threshold
Pending neighborhood, then in the region without scattered points of pending neighborhood, determine insertion point;According to being determined
Insertion point repair described point cloud model.The embodiment of the present invention by travel through point cloud model each scattered points,
In the neighborhood that each scattered points is corresponding, find the scattered points density pending neighborhood less than density threshold, from
And in the region without scattered points of pending neighborhood, determine insertion point, obtain multiple insertion point, passing through gained
The cavity of point cloud model is repaired in the multiple insertion points arrived, and can make the cavity reduction of point cloud model, improve point
The validity of cloud model.
It can be seen that by each scattered points once traveling through point cloud model, it may be determined that go out multiple insertion
Point repairs the cavity of point cloud model so that the cavity reduction of point cloud model, improves the validity of point cloud model;
Therefore optional, the embodiment of the present invention can repeatedly travel through each scattered points of point cloud model, is determined by
Repeatedly the mode of insertion point repairs the cavity of point cloud model so that point cloud model tends to without cavity, further
The validity of hoist point cloud model, wherein, last traversal according to the point cloud model next time traveled through
The point cloud model that insertion point determined by each scattered points of point cloud model is repaired;Traversal number of times can basis
Practical situations sets.
Another flow chart of the point cloud model restorative procedure that Fig. 3 provides for the embodiment of the present invention, with reference to Fig. 3,
The method may include that
Step S200, the number of times of current each scattered points traveling through point cloud model of judgement, if more than setting
Determine number of times, if it is not, perform step S210, if so, perform step S250;
Step S210, will determined by each scattered points of the point cloud model that travel through according to the last time insert
The point cloud model that point is repaired, as point cloud model to be repaired;
Optionally, repair according to insertion point determined by each scattered points of last time traversal point cloud model
The process of point cloud model is referred to shown in Fig. 2, even the upper point cloud model once traveled through each dissipate
In the neighborhood that disorderly point is corresponding, when there is the pending neighborhood that scattered points density is less than density threshold, then treating
Process neighborhood region without scattered points in determine insertion point, according to determined by insertion point carry out point cloud model
Repair;Its essence processing procedure is consistent with Fig. 2;
Optionally, if current point cloud model is initial traverse, the most there is not last traversal point cloud model
The situation of each scattered points, can perform by current point cloud model directly as point cloud model to be repaired
Follow-up flow process.
Step S220, travel through each scattered points of point cloud model to be repaired, and determine traveled through each
The neighborhood that individual scattered points is corresponding;
If in the neighborhood that each scattered points that step S230 is traveled through is corresponding, there is scattered points density low
In the pending neighborhood of density threshold, then in the region without scattered points of pending neighborhood, determine insertion point;
Point cloud model to be repaired is repaired in insertion point determined by step S240, basis;
Step S250, end flow process.
It should be noted that the embodiment of the present invention, in the repair process carrying out point cloud model, is not to lead to
Cross and find the hole region of point cloud model to repair cavity, but by each neighborhood of a point at random of point cloud model
Density carries out the determination of insertion point, thus in the scattered points distribution sparse region interpolation access point of point cloud model,
By the cavity reparation of neighborhood of a point density domination point cloud model at random;Compare the cavity finding point cloud model
The mode in cavity is repaired in region, and the present invention need not find the hole region of point cloud model, computationally intensive
Width reduces, and reparation accuracy is higher, and (the searching amount of calculation of the hole region of point cloud model is very big, and relates to
And decision logic extremely complex, confined condition is more, the wrongheaded situation of hole region easily occurs).
Optionally, really fix exactness for improving the interpolation access point of region without scattered points of pending neighborhood, this
Inventive embodiments can set the size without scattered points region, and concrete can set area as without dissipating
Disorderly put the size in region;Corresponding, it is that pending neighborhood does not exist scattered points without scattered points region
The region setting area, set area less than the area of pending neighborhood;Nothing at pending neighborhood dissipates
Unrest determines in putting region that the process of insertion point can be to there is not the setting of scattered points in pending neighborhood
Insertion point is determined in the region of area.
Preferably, setting area can be with selected point cloud model in the case of existing cavity, a scattered points
Average occupied area, the visual practical situations of concrete numerical value sets.
Fig. 4 shows the setting face that there is not scattered points in pending neighborhood that the embodiment of the present invention provides
Determine the method flow diagram of insertion point in long-pending region, the method mainly according to scattered points each in neighborhood, with
The position relationship of the corresponding scattered points traveled through, the position of constraint insertion point so that determined by insert
Point is positioned at without scattered points region;With reference to Fig. 4, the method may include that
Step S300, determine each scattered points of pending neighborhood, in the corresponding scattered points traveled through
Point set;
Step S310, each point for midpoint set, choose each to pending neighborhood of point of midpoint set
The distance of scattered points meets a distance of first condition, obtains meeting the distance set of first condition;
The embodiment of the present invention can from Point Set close a little to the distance of each scattered points of pending neighborhood,
Selecting the distance meeting first condition, each point of alignment set is all made this and is processed, then can be accorded with
Close the distance set of first condition;
Optionally, the distance meeting first condition can be minimum range, concrete, the embodiment of the present invention
Can choose Point Set close point to the distance minima of each scattered points of pending neighborhood, obtain minimum range
Set;Obviously, first condition can also be the second minimum range, or the 3rd minimum range etc., mainly with
The setting sequence of minimum range is as the determination strategy of first condition.
Step S320, the distance set of first condition meets second condition one will be met apart from corresponding
Midpoint set point be defined as insertion point.
Optionally, the distance meeting second condition can be ultimate range, concrete, the embodiment of the present invention
The point that midpoint corresponding to distance maximum in the distance set meeting first condition is gathered can be defined as
Insertion point;Obviously, second condition can also be the second ultimate range, or the 3rd ultimate range etc., mainly
Sort as the determination strategy of second condition using the setting of ultimate range.
Optionally, insertion point is determined in there is not the region setting area of scattered points in pending neighborhood
Mode can be: determine each scattered points of pending neighborhood, in the corresponding scattered points traveled through
Point set;For each point of midpoint set, determine the point gathered at the midpoint each scattered points to pending neighborhood
Distance minima, obtain minimum range set;Corresponding to the distance maximum in minimum range set
Midpoint set point be defined as insertion point.As example, Fig. 5 shows that the insertion point of correspondence determines and shows
It is intended to.
With reference to Fig. 5, by neighborhood be shaped as spherical as a example by, for ease of example, Fig. 5 represents ball with circular flat
Shape situation;Real point v is the scattered points traveled through, the spheric neighbo(u)rhood of the setting radius size centered by a v
Inside there are other real points in addition to a v in multiple scattered points N (v), i.e. figure;Dissipating in spheric neighbo(u)rhood can be drawn
Disorderly in some N (v), the set Mid (v, N (v)) at the midpoint of each point-to-point v at random, as shown in figure hollow core point;
For each hollow dots, calculate hollow dots to the distance minima of scattered points N (v) in spheric neighbo(u)rhood, obtain
Minimum range set;Using the hollow dots corresponding to the distance maximum in minimum range set as insertion point,
I.e. hollow dots m is the insertion point finally determined.
Optionally, after there is not the region setting area of scattered points in determining pending neighborhood, can be by
Pending neighborhood does not exist the central point in the region setting area of scattered points as insertion point, it is also possible to
It is random in the region setting area that there is not scattered points, to determine insertion point.
Preferably, insertion point is determined in there is not the region setting area of scattered points in pending neighborhood
Mode, can by pending neighborhood is set area region divide realize, searching out not
After there is the zoning of scattered points, then within this does not exist the zoning of scattered points, determine insertion point.
Fig. 6 shows the setting face that there is not scattered points in pending neighborhood that the embodiment of the present invention provides
Determining the other method flow chart of insertion point in long-pending region, with reference to Fig. 6, the method may include that
Step S400, pending neighborhood is divided setting area, obtain multiple there is setting area
Zoning;
With reference to Fig. 7, by neighborhood be shaped as spherical as a example by, for ease of example, Fig. 7 represents ball with circular flat
Shape situation;Real point v is the scattered points traveled through, the spheric neighbo(u)rhood of the setting radius size centered by a v
Inside there is multiple scattered points N (v);The embodiment of the present invention can set area by the single area size divided,
Neighborhood is carried out region division, such as Fig. 7, show that multiple zoning, the size of each zoning are
Set area.
If there is the zoning of scattered points in step S410, then in the division that there is not scattered points
Insertion point is determined in region.
For there is the zoning of scattered points, then the embodiment of the present invention is negligible, does not make insertion point really
Fixed;For there is not the zoning of scattered points, this can not existed the center of the zoning of scattered points,
Or the point randomly selected is defined as insertion point;Optionally, the zoning that there is not scattered points may have many
Individual, the embodiment of the present invention can only determined by a zoning that there is not scattered points determines insertion
Point, obtains an insertion point;Obviously, it is possible in multiple zonings that there is not scattered points the most really
Determine insertion point, obtain multiple insertion point.
With reference to Fig. 7, after neighborhood is carried out region division, dash area is the dividing regions that there is not scattered points
Territory, then can determine insertion point, Diamond spot as shown in Figure 7 in the zoning that there is not scattered points.
Foregoing illustrates the multiple mode determining insertion point in the region without scattered points of pending neighborhood, value
It is noted that mode illustrated above is only optional mode, for other by neighborhood density domination plan
Slightly, in the region without scattered points of pending neighborhood, the mode of insertion point is determined, all in the protection of the present invention
In the range of.
Optionally, after determining insertion point in the region without scattered points of pending neighborhood, the embodiment of the present invention
Can according to determined by insertion point repair point cloud model, point cloud model is carried out mend point so that point cloud model
Scattered points increasing number, the validity of hoist point cloud model.
Further, determined by guarantee, insertion point matches with the morphological characteristic of point cloud model, the present invention
By the normal orientation of insertion point, embodiment can judge whether that determined by employing, point cloud model is made by insertion point
Repair.Fig. 8 shows that the side of point cloud model is repaired in insertion point determined by the basis that the embodiment of the present invention provides
Method flow chart, with reference to Fig. 8, the method may include that
Step S500, for determined by each insertion point, determine determined by insert neighborhood of a point;
Optionally, determine the mode inserting neighborhood of a point, can be with the neighbour determining traveled through scattered points above
The mode in territory is similar to, and can refer to;For simplifying calculation process, the embodiment of the present invention can be directly to be determined
Insertion point corresponding traveled through neighborhood of a point at random, as this insertion neighborhood of a point.
Step S510, judge determined by the normal direction of insertion point, with determined by insert the method for neighborhood of a point
To the most identical, if so, perform step S520, if it is not, perform step S530;
Step S520, retain determined by insertion point, will determined by insertion point as described point cloud model
In be newly added scattered points;
By using determined by insertion point as the scattered points in point cloud model, point cloud model can be mended
Point, the scattered points quantity of hoist point cloud model.
Step S530, abandon determined by insertion point.
Providing below a kind of preferably point cloud model restorative procedure flow process, Fig. 9 provides for the embodiment of the present invention
Another flow chart of point cloud model restorative procedure, with reference to Fig. 9, the method may include that
Step S600, the number of times of current each scattered points traveling through point cloud model of judgement, if more than setting
Determine number of times, if it is not, perform step S610, if so, perform step S690;
Step S610, by according to the last time traversal point cloud model each scattered points determined by insertion point
The point cloud model repaired, as point cloud model to be repaired;
If current point cloud model is initial traverse, the most there is not each of last traversal point cloud model
The situation of scattered points, can be straight by current point cloud model (may be the initial point cloud model of object to be repaired)
Connect as point cloud model to be repaired, perform follow-up flow process.
Step S620, travel through each scattered points of point cloud model to be repaired, and determine traveled through each
The neighborhood that individual scattered points is corresponding;
Step 630, judge whether scattered points density less than the pending neighborhood of density threshold, if so,
Perform step S640, if it is not, perform step S690;
Step S640, for each pending neighborhood, determine each scattered points of pending neighborhood, to corresponding
The midpoint set of the scattered points traveled through, for each point of midpoint set, determines that the point gathered at midpoint is to treating
Process the distance minima of each scattered points of neighborhood, obtain minimum range set, by minimum range set
The midpoint corresponding to distance maximum set point be defined as insertion point;
Wherein, a pending neighborhood can determine that an insertion point.
Step S650, for determined by each insertion point, determine determined by insert neighborhood of a point;
Step S660, judge determined by the normal direction of insertion point, with determined by insert the method for neighborhood of a point
To the most identical, if so, perform step S670, if it is not, perform step S680;
Step S670, retain determined by insertion point, will determined by insertion point as described point cloud model
In be newly added scattered points;
Step S680, abandon determined by insertion point;
Step S690, end flow process.
Above-described point cloud model restorative procedure is applicable to carry out point cloud model arbitrarily mending point
Stage.
Further, when point cloud model carries out mending point, the embodiment of the present invention can be divided into the filling-up hole stage and add
Close stage, filling-up hole stage and encrypting stage process logical AND point cloud model restorative procedure described above
Process logic is similar to, and difference is the contiguous range selected by the filling-up hole stage, more than encrypting stage institute
The contiguous range chosen.
The another flow chart of the point cloud model restorative procedure that Figure 10 provides for the embodiment of the present invention, with reference to Figure 10,
The method may include that
Step S700, determine point cloud model to be repaired;
Step S710, travel through each scattered points of point cloud model to be repaired, and determine traveled through each
The filling-up hole neighborhood that individual scattered points is corresponding;
If in the filling-up hole neighborhood that each scattered points that step S720 is traveled through is corresponding, there is scattered points close
Degree is less than the pending filling-up hole neighborhood of density threshold, then in the region without scattered points of pending filling-up hole neighborhood
Determine insertion point;
Insertion point determined by step S730, basis carries out filling-up hole to point cloud model to be repaired;
It should be noted that step S700~step S730 are the filling-up hole stage, the filling-up hole stage determines complex point to be repaired
The mode of cloud model can be: initial point cloud model to be repaired is defined as described point cloud model;Or,
If the number of times having traveled through each scattered points of point cloud model to be repaired is not up to set point number, then by basis
The point cloud model that insertion point determined by each scattered points of the point cloud model of last traversal is repaired, makees
For current point cloud model to be repaired;Wherein, the point cloud model next time traveled through is, according to last time
The point cloud model that the insertion point that each scattered points of the point cloud model gone through determines is repaired.
Step S740, determine point cloud model to be encrypted;
Determined by point cloud model to be encrypted can be the filling-up hole stage, traveled through the every of point cloud model to be repaired
When the number of times of one scattered points reaches set point number, corresponding point cloud model;Or, the filling-up hole stage, institute
When the scattered points density of the filling-up hole neighborhood corresponding to each scattered points of traversal is all not less than density threshold, institute
Corresponding point cloud model.
Step S750, travel through each scattered points of point cloud model to be encrypted, and determine traveled through each
The encryption neighborhood that individual scattered points is corresponding, wherein, the scope of encryption neighborhood is less than the scope of filling-up hole neighborhood;
Optionally, if neighborhood is regular shape, as spherical etc., then the radius encrypting neighborhood need to be less than filling-up hole
The radius of neighborhood.
If in the encryption neighborhood that each scattered points that step S760 is traveled through is corresponding, there is scattered points close
Degree is less than the pending encryption neighborhood of density threshold, then in the region without scattered points of pending encryption neighborhood
Determine insertion point;
Described point cloud model to be encrypted is repaired in insertion point determined by step S770, basis.
It can be seen that step S740~step S760 are encrypting stage, it is that step S700~step S730 are mended
Point cloud model behind hole is encrypted, and the process logic of filling-up hole stage and encrypting stage is basically identical, both
Result be all regarded as point cloud model is carried out mending point;Can refer to the description of Fig. 2~Fig. 9 part above.Mend
The difference of hole stage and encrypting stage is: the contiguous range selected by the filling-up hole stage is relatively big, first bigger
In the range of determine insertion point by neighborhood density domination, with carry out mend point;And the neighbour selected by encrypting stage
Territory scope is less, after the filling-up hole stage, is determined by neighborhood density domination further in less scope
Insertion point, to promote the quantity mending point.
One application of the point cloud model restorative procedure that the embodiment of the present invention provides is, to streetscape point cloud model
In building point cloud model carry out cavity reparation.Streetscape point cloud model mainly by building point cloud model and
Perigee cloud model is constituted, and the embodiment of the present invention can be partitioned into building object point cloud from streetscape point cloud model
After model, the detached building point cloud model being partitioned into is carried out cavity reparation;Before repair mode can refer to
The description of texts and pictures 2~Fig. 9 part, it is clear that the mode that may be used without filling-up hole and encryption combination carries out cavity reparation.
Below in the way of using filling-up hole and encryption combination, the repair process of building point cloud model is carried out
Introduce:
1, from streetscape point cloud model, building point cloud model it is partitioned into, the building object point cloud mould being partitioned into
Type should be independent;Concrete, it may be determined that the building point cloud model in streetscape point cloud model, by difference
The building point cloud model of height is split, and is clustered mutually by mutually level building point cloud model,
To the building point cloud model that at least one is independent;
2, after being partitioned into independent building point cloud model, each of building point cloud model can be traveled through
Scattered points, and determine the filling-up hole neighborhood that each scattered points traveled through is corresponding;Optionally, can set
First radius determines the spherical filling-up hole neighborhood that each scattered points is corresponding, sets the first radius and can choose and build
Build the half of thing story height, it is clear that depending on the concrete the most visual actually used situation of numerical value;
If in the filling-up hole neighborhood that 3 each scattered points traveled through are corresponding, there is scattered points density less than close
The pending filling-up hole neighborhood of degree threshold value, then determine insertion in the region without scattered points of pending filling-up hole neighborhood
Point;Optional determine that mode can be as illustrated in fig. 4 or 6;
4, the normal direction of insertion point is compared with its neighborhood normal direction, retain with determined by the neighbour of insertion point
The insertion point that the normal direction in territory is identical, passes through retained insertion point and carries out filling-up hole;
5, circulation performs above-mentioned 2~4 steps, until each scattered points of traversal building point cloud model
When number of times reaches set point number, or, the filling-up hole neighborhood that each scattered points of being traveled through is corresponding at random
When dot density is all not less than density threshold, enter step 6;It should be noted that the building next time traveled through
Object point cloud model is, according to inserting that each scattered points of the building point cloud model of last time traversal determines
Access point carries out the building point cloud model after filling-up hole;
6, by the building point cloud model after the filling-up hole obtained by execution step 5, as building to be encrypted
Point cloud model;
7, travel through each scattered points of point cloud model to be encrypted, and determine each scattered points traveled through
Corresponding encryption neighborhood, wherein, the scope of encryption neighborhood is less than the scope of filling-up hole neighborhood;Optionally, may be used
To set the spherical encryption neighborhood that the second radius determines that each scattered points is corresponding, set the second radius and be less than
Set the first radius, set the second radius and can choose 0.5 meter, it is clear that specifically numerical value is actual makes
Depending on situation;
If in the encryption neighborhood that 8 each scattered points traveled through are corresponding, there is scattered points density less than close
The pending encryption neighborhood of degree threshold value, then determine insertion in the region without scattered points of pending encryption neighborhood
Point;Optional determine that mode can be as illustrated in fig. 4 or 6;
9, the normal direction of insertion point is compared with its neighborhood normal direction, retain with determined by the neighbour of insertion point
The insertion point that the normal direction in territory is identical, passes through retained insertion point and is encrypted;
10, circulation performs above-mentioned 7~9 steps, until each scattered points of traversal building point cloud model
Number of times when reaching set point number, or, dissipating of the encryption neighborhood that each scattered points of being traveled through is corresponding
When random dot density is all not less than density threshold, the building point cloud model after being repaired;Should be noted that
It is the building point cloud model next time traveled through is, according to the building point cloud model of last time traversal
The insertion point that each scattered points determines be encrypted after building point cloud model;
11, due to perigee cloud model object be distributed especially complex, mainly have trees, pedestrian, vehicle,
The object such as billboard and various traffic signboards;This part data major part all ratio is more complete, Bu Guoyou
The uneven meeting of speed in collecting vehicle operation causes the scattered points skewness of perigee cloud model;Therefore
After step 10, perigee cloud model can be encrypted, improve the quality of streetscape point cloud model;
The process that perigee cloud model is encrypted, similar with the process that building point cloud model is encrypted,
Can phase reference.
The embodiment of the present invention is by traveling through each scattered points of point cloud model, in the neighborhood that each scattered points is corresponding,
Find scattered points density less than the pending neighborhood of density threshold, thus at pending neighborhood without scattered points
Determine insertion point in region, obtain multiple insertion point, repair some cloud mould by obtained multiple insertion points
The cavity of type, can make the cavity reduction of point cloud model, improve the validity of point cloud model;Meanwhile, originally
Invention is repaiied by the cavity of neighborhood of a point density domination point cloud model at random, it is not necessary to find point cloud model
Hole region, amount of calculation significantly reduces, and it is higher to repair accuracy.
The point cloud model prosthetic device provided the embodiment of the present invention below is introduced, point described below
Cloud model prosthetic device can be with above-described point cloud model restorative procedure mutually to should refer to.
The structured flowchart of the point cloud model prosthetic device that Figure 11 provides for the embodiment of the present invention, this device can be answered
For having the calculating equipment that point cloud model can be carried out data process of data-handling capacity;With reference to Figure 11,
This device may include that
Model determines module 100, is used for determining point cloud model to be repaired;
Neighborhood determines module 200, for traveling through each scattered points of described point cloud model, and determines institute time
The neighborhood that each scattered points of going through is corresponding;
Insertion point determines module 300, if in the neighborhood that each scattered points of being traveled through is corresponding, existing
Scattered points density is less than the pending neighborhood of density threshold, then in the region without scattered points of pending neighborhood
Determine insertion point;
Repair module 400, repairs described point cloud model for insertion point determined by basis.
Optionally, without scattered points region can be the setting area that there is not scattered points in pending neighborhood
Region;Described setting area is less than the area of pending neighborhood;Figure 12 shows that the embodiment of the present invention provides
Insertion point determine the optional structure of one of module 300, with reference to Figure 12, insertion point determines that module 300 can
To include:
Insertion point determines performance element 310, for there is not the setting area of scattered points in pending neighborhood
Region in determine insertion point.
Optionally, Figure 13 shows that the insertion point that the embodiment of the present invention provides determines the one of performance element 310
Planting optional structure, with reference to Figure 13, insertion point determines that performance element 310 may include that
Midpoint set determines subelement 3101, for determining each scattered points of pending neighborhood, to correspondence
The midpoint set of the scattered points traveled through;
Choosing subelement 3102, for each point for midpoint set, the point choosing midpoint set is located to waiting
The distance of each scattered points of reason neighborhood meets a distance of first condition, obtain meeting first condition away from
From set;
Optionally, the distance meeting first condition can be minimum range.
Subelement 3103 is chosen in insertion point, for meeting Article 2 by meeting in the distance set of first condition
The one of part is defined as insertion point apart from the point of corresponding midpoint set;
Optionally, the distance meeting second condition can be ultimate range.
Optionally, Figure 14 shows that the insertion point that the embodiment of the present invention provides determines the another of performance element 310
A kind of optional structure, with reference to Figure 14, insertion point determines that performance element 310 may include that
Divide subelement 3111, for pending neighborhood being divided with setting area, obtain multiple tool
There is the zoning setting area;
Subelement 3112 is determined, if for the zoning that there is scattered points, then in zoning
Insertion point is determined in the zoning that there is not scattered points.
Optionally, Figure 15 shows the optional structure of one of the repair module 400 that the embodiment of the present invention provides,
With reference to Figure 15, repair module 400 may include that
Insertion point neighborhood determines unit 410, for for determined by each insertion point, determine determined by insert
Enter neighborhood of a point;
Normal direction judging unit 420, the normal direction of insertion point determined by judge, with determined by insertion point
The normal direction of neighborhood the most identical;
Stick unit 430, if the normal direction of insertion point determined by Yong Yu, with determined by insert neighborhood of a point
Normal direction identical, then retain determined by insertion point, will determined by insertion point as described point cloud model
In be newly added scattered points;
Discarding unit 440, if the normal direction of insertion point determined by Yong Yu, with determined by insert neighborhood of a point
Normal direction different, then abandon determined by insertion point.
Optionally, model determines that module 100 determines that the process of point cloud model to be repaired can be concrete: if
The number of times of each scattered points of current traversal point cloud model is not up to set point number, then will be according to the last time
The point cloud model that insertion point determined by each scattered points of traversal point cloud model is repaired, as to be repaired
Point cloud model.Obviously, if current point cloud model is initial traverse, the most there is not last traversal point cloud
The situation of each scattered points of model, can be by current point cloud model directly as point cloud model to be repaired.
Optionally, the embodiment of the present invention can use the mode of filling-up hole combining encryption, repaiies point cloud model
Multiple.Functional module shown in Figure 11 can perform the filling-up hole stage;Concrete, repair module 400 can be according to institute
The insertion point determined carries out filling-up hole to described point cloud model, and correspondence, neighborhood is filling-up hole neighborhood;Model is true
Point cloud model to be repaired determined by cover half block 100 can be, initial point cloud model to be repaired, or,
When the number of times of each scattered points having traveled through point cloud model to be repaired is not up to set point number, according to upper one
The point cloud model that insertion point determined by each scattered points of the point cloud model of secondary traversal is repaired;Wherein,
The point cloud model next time traveled through is, determines according to each scattered points of the point cloud model of last time traversal
The point cloud model repaired of insertion point.On this basis, Figure 16 shows what the embodiment of the present invention provided
Another structured flowchart of point cloud model prosthetic device, with reference to Figure 16, this device can also include:
Point cloud model to be encrypted determines module 500, is used for determining point cloud model to be encrypted;
Optionally, point cloud model to be encrypted can be in the stage that described point cloud model carries out filling-up hole,
When the number of times of each scattered points traveling through point cloud model to be repaired reaches set point number, corresponding some cloud
Model, or, in the stage that described point cloud model carries out filling-up hole, each scattered points traveled through is corresponding
The scattered points density of filling-up hole neighborhood when being all not less than density threshold, corresponding point cloud model.
Encryption neighborhood determines module 600, for traveling through each scattered points of point cloud model to be encrypted, and really
The encryption neighborhood that fixed each scattered points traveled through is corresponding, wherein, the scope of encryption neighborhood is less than filling-up hole
The scope of neighborhood
Encryption insertion point determines module 700, if the encryption neighborhood that each scattered points for being traveled through is corresponding
In, there is the scattered points density pending encryption neighborhood less than density threshold, then at pending encryption neighborhood
Region without scattered points in determine insertion point;
Encryption repair module 800, repairs described point cloud model to be encrypted for insertion point determined by basis.
Optionally, encryption insertion point determines that the structure of module 700 can determine the knot of module 300 with insertion point
Structure is similar to, can phase reference;The structure of encryption repair module 800 can be with repair module 400 shown in Figure 15
Structure is similar to, can phase reference.
Optionally, the point cloud model prosthetic device that the embodiment of the present invention provides can be applicable to streetscape point cloud mould
Building point cloud model in type carries out cavity reparation;The embodiment of the present invention can building point cloud model by
After streetscape point cloud model is partitioned into, independent building point cloud model is carried out cavity reparation.Figure 17 shows
Go out another structured flowchart of the point cloud model prosthetic device that the embodiment of the present invention provides, in conjunction with Figure 11 and Tu
Shown in 17, this device can also include:
Segmentation module 900, for determining the building point cloud model in streetscape point cloud model, by differing heights
Building point cloud model split, mutually level building point cloud model is clustered mutually, obtain to
A few building point cloud model;
Obtained building point cloud model can carry out cavity by point cloud model prosthetic device shown in Figure 11 and repair
Multiple, it is possible to by point cloud model prosthetic device shown in Figure 16, to use the mode of filling-up hole combining encryption.
Further, after repairing building point cloud model, the embodiment of the present invention also can be to perigee
Cloud model is encrypted.
The embodiment of the present invention also provides for a kind of calculating equipment, and this calculating equipment can include point described above
Cloud model prosthetic device;This calculating equipment, can be by traversal point cloud model when carrying out point cloud model and repairing
Each scattered points, in the neighborhood that each scattered points is corresponding, find scattered points density less than the treating of density threshold
Process neighborhood, thus determine insertion point in the region without scattered points of pending neighborhood, obtain multiple insertion
Point, is repaired the cavity of point cloud model, can make the cavity of point cloud model by obtained multiple insertion points
Reduce, improve the validity of point cloud model;Meanwhile, this calculating equipment is by neighborhood of a point density control at random
The cavity of point cloud model processed is repaiied, it is not necessary to finding the hole region of point cloud model, amount of calculation significantly reduces,
And it is higher to repair accuracy.
The hardware block diagram calculating equipment that Figure 18 provides for the embodiment of the present invention, the equipment of calculating can be as
The subscriber equipmenies such as smart mobile phone, PC (personal computer), or the network equipment such as server;With reference to figure
18, calculate and may include that processor 1, communication interface 2, memorizer 3 and communication bus 4;
Wherein processor 1, communication interface 2, memorizer 3 complete mutual communication by communication bus 4;
Optionally, communication interface 2 can be the interface of communication module, such as the interface of gsm module;
Processor 1, is used for performing program;
Memorizer 3, is used for depositing program;
Program can include that program code, described program code include computer-managed instruction.
Processor 1 is probably a central processor CPU, or specific integrated circuit ASIC
(Application Specific Integrated Circuit), or be configured to implement the embodiment of the present invention
One or more integrated circuits.
Memorizer 3 may comprise high-speed RAM memorizer, it is also possible to also includes nonvolatile memory
(non-volatile memory), for example, at least one disk memory.
Wherein, program can be specifically for:
Determine point cloud model to be repaired;
Travel through each scattered points of described point cloud model, and determine that each scattered points traveled through is corresponding
Neighborhood;
If in the neighborhood that each scattered points traveled through is corresponding, there is scattered points density less than density threshold
Pending neighborhood, then in the region without scattered points of pending neighborhood, determine insertion point;
Described point cloud model is repaired in insertion point determined by according to.
In this specification, each embodiment uses the mode gone forward one by one to describe, and each embodiment stresses
Being the difference with other embodiments, between each embodiment, identical similar portion sees mutually.
For device disclosed in embodiment, owing to it corresponds to the method disclosed in Example, so describing
Fairly simple, relevant part sees method part and illustrates.
Professional further appreciates that, respectively shows in conjunction with what the embodiments described herein described
The unit of example and algorithm steps, it is possible to electronic hardware, computer software or the two be implemented in combination in,
In order to clearly demonstrate the interchangeability of hardware and software, the most general according to function
Describe composition and the step of each example.These functions perform with hardware or software mode actually,
Depend on application-specific and the design constraint of technical scheme.Professional and technical personnel can be to each specific
Should be used for use different methods to realize described function, but this realization is it is not considered that beyond this
The scope of invention.
The method described in conjunction with the embodiments described herein or the step of algorithm can directly use hardware,
The software module that processor performs, or the combination of the two implements.Software module can be placed in and deposit at random
Reservoir (RAM), internal memory, read only memory (ROM), electrically programmable ROM, electric erasable can
Well known in programming ROM, depositor, hard disk, moveable magnetic disc, CD-ROM or technical field
In any other form of storage medium.
Described above to the disclosed embodiments, makes professional and technical personnel in the field be capable of or uses
The present invention.Multiple amendment to these embodiments will be aobvious and easy for those skilled in the art
See, generic principles defined herein can without departing from the spirit or scope of the present invention,
Realize in other embodiments.Therefore, the present invention is not intended to be limited to the embodiments shown herein,
And it is to fit to the widest scope consistent with principles disclosed herein and features of novelty.
Claims (15)
1. a point cloud model restorative procedure, it is characterised in that including:
Determine point cloud model to be repaired;
Travel through each scattered points of described point cloud model, and determine that each scattered points traveled through is corresponding
Neighborhood;
If in the neighborhood that each scattered points traveled through is corresponding, there is scattered points density less than density threshold
Pending neighborhood, then in the region without scattered points of pending neighborhood, determine insertion point;
Described point cloud model is repaired in insertion point determined by according to.
Point cloud model restorative procedure the most according to claim 1, it is characterised in that described without at random
Point region is the region setting area that there is not scattered points in pending neighborhood;Described setting area is less than
The area of pending neighborhood;
Described in the region without scattered points of pending neighborhood, determine that insertion point includes:
Insertion point is determined in there is not the region setting area of scattered points in pending neighborhood.
Point cloud model restorative procedure the most according to claim 2, it is characterised in that described locate waiting
Determine that insertion point includes in reason neighborhood does not exist the region setting area of scattered points:
Determine each scattered points of pending neighborhood, to the midpoint set of the corresponding scattered points traveled through;
For midpoint set each point, choose midpoint set point to pending neighborhood each scattered points away from
Meet a distance of first condition in from, obtain meeting the distance set of first condition;
The distance set of first condition meets second condition one will be met apart from corresponding midpoint set
Point be defined as insertion point.
Point cloud model restorative procedure the most according to claim 2, it is characterised in that described locate waiting
Determine that insertion point includes in reason neighborhood does not exist the region setting area of scattered points:
With setting area, pending neighborhood is divided, obtain multiple zoning having and setting area;
If there is the zoning of scattered points, then determine in the zoning that there is not scattered points
Insertion point.
5. according to the point cloud model restorative procedure described in any one of claim 1-4, it is characterised in that institute
State the insertion point described point cloud model of reparation determined by basis to include:
Each insertion point determined by for, determine determined by insert neighborhood of a point;
The normal direction of insertion point determined by judgement, with determined by insert the normal direction of neighborhood of a point the most identical;
The normal direction of insertion point determined by if, with determined by insert the normal direction of neighborhood of a point identical, then protect
Insertion point determined by staying, using determined by insertion point as described point cloud model is newly added scattered points;
The normal direction of insertion point determined by if, from determined by insert the normal direction of neighborhood of a point different, then lose
Insertion point determined by abandoning.
Point cloud model restorative procedure the most according to claim 1, it is characterised in that described determine treat
Reparation point cloud model includes:
If currently the number of times of each scattered points of traversal point cloud model is not up to set point number, then by basis
The point cloud model that insertion point determined by each scattered points of last traversal point cloud model is repaired, as
Point cloud model to be repaired.
Point cloud model restorative procedure the most according to claim 1, it is characterised in that described according to institute
The insertion point determined is repaired described point cloud model and is included:
Insertion point determined by according to carries out filling-up hole to described point cloud model, and wherein, described neighborhood is filling-up hole
Neighborhood;
Described determine that point cloud model to be repaired includes:
Initial point cloud model to be repaired is defined as current point cloud model to be repaired;
Or, if the number of times having traveled through each scattered points of point cloud model to be repaired is not up to set point number,
The point cloud then insertion point determined by each scattered points of the point cloud model traveled through according to the last time repaired
Model, as current point cloud model to be repaired;Wherein, the point cloud model next time traveled through is, according to
The point cloud model that the insertion point that each scattered points of the point cloud model of last traversal determines is repaired;
Described method also includes:
Determine point cloud model to be encrypted;Wherein, described point cloud model to be encrypted is to described point cloud model
Carrying out the stage of filling-up hole, the number of times of each scattered points having traveled through point cloud model to be repaired reaches setting time
During number, corresponding point cloud model, or, in the stage that described point cloud model is carried out filling-up hole, traveled through
The scattered points density of filling-up hole neighborhood corresponding to each scattered points when being all not less than density threshold, corresponding
Point cloud model;
Travel through each scattered points of point cloud model to be encrypted, and determine each scattered points pair traveled through
The encryption neighborhood answered, wherein, the scope of encryption neighborhood is less than the scope of filling-up hole neighborhood;
If in the encryption neighborhood that each scattered points traveled through is corresponding, there is scattered points density less than density
The pending encryption neighborhood of threshold value, then determine insertion point in the region without scattered points of pending encryption neighborhood;
Described point cloud model to be encrypted is repaired in insertion point determined by according to.
Point cloud model restorative procedure the most according to claim 1, it is characterised in that described some cloud mould
Type is building point cloud model;Described building point cloud model is partitioned into by streetscape point cloud model, described
Streetscape point cloud model includes: building point cloud model and perigee cloud model;
Described included by the process being partitioned into building point cloud model in streetscape point cloud model:
Determine the building point cloud model in streetscape point cloud model;
The building point cloud model of differing heights is split, by mutually level building point cloud model
Cluster mutually, obtain at least one building point cloud model.
9. a point cloud model prosthetic device, it is characterised in that including:
Model determines module, is used for determining point cloud model to be repaired;
Neighborhood determines module, for traveling through each scattered points of described point cloud model, and determines and is traveled through
Neighborhood corresponding to each scattered points;
Insertion point determines module, if in the neighborhood that each scattered points of being traveled through is corresponding, existing scattered
Random dot density is less than the pending neighborhood of density threshold, then in the region without scattered points of pending neighborhood really
Determine insertion point;
Repair module, repairs described point cloud model for insertion point determined by basis.
Point cloud model prosthetic device the most according to claim 9, it is characterised in that described nothing dissipates
Disorderly some region is the region setting area that there is not scattered points in pending neighborhood;Described setting area is little
Area in pending neighborhood;Described insertion point determines that module includes:
Insertion point determines performance element, for there is not the setting area of scattered points in pending neighborhood
Insertion point is determined in region.
11. point cloud model prosthetic devices according to claim 10, it is characterised in that described insertion
Point determines that performance element includes:
Midpoint set determines subelement, for determining each scattered points of pending neighborhood, to corresponding institute time
The midpoint set of the scattered points gone through;
Choose subelement, for each point for midpoint set, choose the point of midpoint set to pending neighbour
The distance of each scattered points in territory meets a distance of first condition, obtains meeting the distance set of first condition
Close;
Subelement is chosen in insertion point, for meeting second condition by meeting in the distance set of first condition
Point gathered apart from corresponding midpoint is defined as insertion point.
12. point cloud model prosthetic devices according to claim 10, it is characterised in that described insertion
Point determines that performance element includes:
Divide subelement, for set area pending neighborhood is divided, obtain multiple have set
Determine the zoning of area;
Subelement is determined, if for the zoning that there is scattered points, then not in zoning
Insertion point is determined in there is the zoning of scattered points.
13. according to the point cloud model prosthetic device described in any one of claim 9-12, it is characterised in that
Described repair module includes:
Insertion point neighborhood determines unit, for for determined by each insertion point, determine determined by insert
Neighborhood of a point;
Normal direction judging unit, determined by the judge normal direction of insertion point, with determined by insertion point
The normal direction of neighborhood is the most identical;
Stick unit, if the normal direction of insertion point determined by Yong Yu, with determined by insert neighborhood of a point
Normal direction is identical, then insertion point determined by reservation, and determined by general, insertion point is as in described point cloud model
Be newly added scattered points;
Discarding unit, if the normal direction of insertion point determined by Yong Yu, with determined by insert neighborhood of a point
Normal direction is different, then abandon determined by insertion point.
14. point cloud model prosthetic devices according to claim 9, it is characterised in that described reparation
Module carries out filling-up hole specifically for insertion point determined by basis to described point cloud model;Described neighborhood is for mending
Hole neighborhood;
Described model determines that module is current to be repaired specifically for being defined as by initial point cloud model to be repaired
Complex point cloud model;Or, if the number of times having traveled through each scattered points of point cloud model to be repaired not up to sets
Determine number of times, then insertion point determined by each scattered points of the point cloud model traveled through according to the last time is repaiied
Multiple point cloud model, as current point cloud model to be repaired;Wherein, the point cloud model next time traveled through
For, the some cloud that the insertion point determined according to each scattered points of the point cloud model of last time traversal is repaired
Model;
Described point cloud model prosthetic device also includes:
Point cloud model to be encrypted determines module, is used for determining point cloud model to be encrypted;Wherein, described to be added
Close point cloud model is in the stage that described point cloud model carries out filling-up hole, has traveled through point cloud model to be repaired
When the number of times of each scattered points reaches set point number, corresponding point cloud model, or, to described point
Cloud model carries out the stage of filling-up hole, and the scattered points of the filling-up hole neighborhood that each scattered points of being traveled through is corresponding is close
When degree is all not less than density threshold, corresponding point cloud model;
Encryption neighborhood determines module, for traveling through each scattered points of point cloud model to be encrypted, and determines
The encryption neighborhood that each scattered points of being traveled through is corresponding, wherein, the scope of encryption neighborhood is adjacent less than filling-up hole
The scope in territory
Encryption insertion point determines module, if in the encryption neighborhood that each scattered points of being traveled through is corresponding,
There is the scattered points density pending encryption neighborhood less than density threshold, then in the nothing of pending encryption neighborhood
Scattered points determines insertion point in region;
Encryption repair module, repairs described point cloud model to be encrypted for insertion point determined by basis.
15. 1 kinds calculate equipment, it is characterised in that include the some cloud described in any one of claim 9-14
Model prosthetic device.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201510111559.6A CN106033620B (en) | 2015-03-13 | 2015-03-13 | A kind of point cloud model restorative procedure, device and computing device |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201510111559.6A CN106033620B (en) | 2015-03-13 | 2015-03-13 | A kind of point cloud model restorative procedure, device and computing device |
Publications (2)
Publication Number | Publication Date |
---|---|
CN106033620A true CN106033620A (en) | 2016-10-19 |
CN106033620B CN106033620B (en) | 2018-10-19 |
Family
ID=57150602
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201510111559.6A Active CN106033620B (en) | 2015-03-13 | 2015-03-13 | A kind of point cloud model restorative procedure, device and computing device |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106033620B (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112287907A (en) * | 2020-12-24 | 2021-01-29 | 湖南联智科技股份有限公司 | Hole identification method based on point cloud density gradient |
CN112767554A (en) * | 2021-04-12 | 2021-05-07 | 腾讯科技(深圳)有限公司 | Point cloud completion method, device, equipment and storage medium |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101127123A (en) * | 2007-09-11 | 2008-02-20 | 东南大学 | Sign point hole filling method based on neural network in tri-D scanning point cloud |
US20130321393A1 (en) * | 2012-05-31 | 2013-12-05 | Microsoft Corporation | Smoothing and robust normal estimation for 3d point clouds |
US20140324405A1 (en) * | 2013-03-15 | 2014-10-30 | State Farm Mutual Automobile Insurance Company | Systems and Methods for Assessing a Roof and Generating Models |
CN104282039A (en) * | 2014-09-29 | 2015-01-14 | 樊晓莉 | Skeleton orthosis brace shaping method based on 3D scanning |
-
2015
- 2015-03-13 CN CN201510111559.6A patent/CN106033620B/en active Active
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101127123A (en) * | 2007-09-11 | 2008-02-20 | 东南大学 | Sign point hole filling method based on neural network in tri-D scanning point cloud |
US20130321393A1 (en) * | 2012-05-31 | 2013-12-05 | Microsoft Corporation | Smoothing and robust normal estimation for 3d point clouds |
US20140324405A1 (en) * | 2013-03-15 | 2014-10-30 | State Farm Mutual Automobile Insurance Company | Systems and Methods for Assessing a Roof and Generating Models |
CN104282039A (en) * | 2014-09-29 | 2015-01-14 | 樊晓莉 | Skeleton orthosis brace shaping method based on 3D scanning |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112287907A (en) * | 2020-12-24 | 2021-01-29 | 湖南联智科技股份有限公司 | Hole identification method based on point cloud density gradient |
CN112767554A (en) * | 2021-04-12 | 2021-05-07 | 腾讯科技(深圳)有限公司 | Point cloud completion method, device, equipment and storage medium |
Also Published As
Publication number | Publication date |
---|---|
CN106033620B (en) | 2018-10-19 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN112465948B (en) | Vehicle-mounted laser pavement point cloud rarefying method capable of retaining spatial features | |
CN110906940B (en) | Lane sideline aggregation method based on track direction | |
CN112154446B (en) | Stereo lane line determining method and device and electronic equipment | |
CN106548520A (en) | A kind of method and system of cloud data denoising | |
CN108709513A (en) | A kind of three-dimensional vehicle detection method based on model-fitting algorithms | |
CN102750730B (en) | Characteristic-maintained point cloud data compacting method | |
CN107705363B (en) | Road three-dimensional visual modeling method and device | |
CN110598541A (en) | Method and equipment for extracting road edge information | |
CN111897906A (en) | Method, device, equipment and storage medium for processing map data | |
CN105550688A (en) | Method and device for point cloud data classification | |
CN111651712A (en) | Method and system for evaluating complexity of test scene of intelligent automobile | |
CN106033620A (en) | Point clouds model restoration method, point clouds model restoration device and calculation device | |
CN102800114B (en) | Data point cloud downsizing method based on Poisson-disk sampling | |
CN113836484B (en) | Self-adaptive point cloud thinning method based on path point adjacent domain and ground filtering | |
CN115346183A (en) | Lane line detection method, terminal and storage medium | |
CN110060266B (en) | Lane line extraction method and apparatus, server, and computer-readable medium | |
EP0644497B1 (en) | Method and system for dividing analyzing region in device simulator | |
CN116764225A (en) | Efficient path-finding processing method, device, equipment and medium | |
CN116246069A (en) | Method and device for self-adaptive terrain point cloud filtering, intelligent terminal and storage medium | |
CN113239906B (en) | Lane line detection method and device | |
CN113239905A (en) | Lane line simplification method and device, electronic equipment and storage medium | |
CN115861566A (en) | LiDAR point cloud building integrity enhancement method based on intensity voxel model | |
CN110887503B (en) | Moving track simulation method, device, equipment and medium | |
JP2018124687A (en) | Probe information processing device, probe information processing method, computer program, and recording medium for recording computer program | |
CN113052231A (en) | Three-dimensional model classification method based on voxel and global shape distribution characteristics |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant | ||
TR01 | Transfer of patent right | ||
TR01 | Transfer of patent right |
Effective date of registration: 20230706 Address after: 518000 Tencent Building, No. 1 High-tech Zone, Nanshan District, Shenzhen City, Guangdong Province, 35 Floors Patentee after: TENCENT TECHNOLOGY (SHENZHEN) Co.,Ltd. Patentee after: TENCENT CLOUD COMPUTING (BEIJING) Co.,Ltd. Address before: 2, 518000, East 403 room, SEG science and Technology Park, Zhenxing Road, Shenzhen, Guangdong, Futian District Patentee before: TENCENT TECHNOLOGY (SHENZHEN) Co.,Ltd. |